Prior research has found that external audits of banks under the Federal Deposit Insurance Corporation Improvement Act (‘FDICIA’) of 1991 constrained at least some aspects of accounting discretion. This is important, in part, because subsequent regulation, including the Sarbanes-Oxley Act (‘SOX’) of 2002, imposed similar provisions, including those requiring independent audit committees and attestation of internal controls. The purpose was to improve corporate governance with high quality financial reporting.

The methodologies of these studies, however, have been criticized for failing to isolate discretionary and non-discretionary components of accounting decisions, and for failing to disentangle differences in accounting behaviors due to audit regulation from those due to omitted variables that are correlated with bank size. In our paper 'The Reach and Effectiveness of FDICIA: Internal Controls and Audit Committee Regulation in Banking', we examine accounting discretion, alternatively, using confidential supervisory data to establish two separate metrics that do not rely on a breakdown of components of accounting decisions, and are equally applicable across banks of all sizes.

One metric is a continuous probability of a bank experiencing a downgrade in its overall supervisory rating. Although this (numeric) rating for a given bank is established during on-site examinations, when the information analyzed by regulators is both public and private, the probability of its downgrade is generated off-site, in periods between on-site examinations, using information from publicly-reported financial statements only. We hypothesize that the usefulness of financial statements in predicting downgrades will increase when they are conditioned by audit under internal controls and audit committee requirements. The auditing reforms of FDICIA were intended specifically to do this, ie, to render financial statements more ‘reliable’ in predicting ‘deterioration’ in financial condition.

Using a sample of 45,633 annual observations on commercial banks, from 2005 to 2015, we find that the financial statements of banks subject to audits encompassing internal controls and audit committee requirements of FDICIA, both independently and jointly with similar requirements under SOX, have greater capacity to predict subsequent downgrades in supervisory ratings than banks that are audited in the absence of internal controls and audit committee regulation. The benefits of audit under SOX alone, however, are not as readily apparent as those of audits of banks subject to FDICIA. And the findings of an effect of audits under FDICIA are not uniform across time.

The second metric uses confidential supervisory data on regulatory ratings of operational policies that incorporate accounting practices. We find that regulators assign better absolute ratings to banks audited under internal controls and audit committee regulations of FDICIA. After incorporating the impacts of inherent accounting risk, however, the differentials diminish and are apparent only to a more limited extent in relative ratings. Collectively, our evidence suggests that the internal controls and audit committee provisions of FDICIA may have value in monitoring accounting practices among the dwindling numbers of banks that remain subject to them and to them alone (a larger percentage of bank assets are currently covered by internal controls and audit committee regulation under SOX than was covered under FDICIA at its inception). From a more general perspective of accounting practices, it offers a unique, if qualified, confirmation of an impact of auditors on financial reporting.

Drew Dahl is an economist at the Federal Reserve Bank of St. Louis. Richard Jenson is a professor in the School of Accountancy at Utah State University. The views expressed are those of the individual authors and do not necessarily reflect official positions of the Federal Reserve Bank of St. Louis, the Federal Reserve System, or the Board of Governors.